Temperature and Molecular Attraction. 109 



(The constant in this equation may not be the same constant 

 as in equation 6, but there is no way of comparing the 

 absolute values of the constants. Otherwise the equations 

 are identical.) I have as yet found no satisfactory way of 

 determining the "mean orbital"" energy retained by the n 

 molecules in coming together, but it seems certain from the 

 constancy of the results shown in Table I. that the amount 

 of orbital energy retained is either zero or proportional to 

 the amount of energy given out. (In my earlier papers I 

 expressed the opinion that the total kinetic energy of a mole- 

 cule in the liquid and in the gaseous condition were the same. 

 This statement I am now convinced was an error. The 

 liquid molecules retain, in addition to their temperature 

 energy, an " orbital " energy, which being proportional to 

 the energy given out, that is to the internal heat of vapori- 

 zation, was at the time undetected.) 



Although certainty is impossible, it would seem pretty 

 safe to guess that when n bodies are drawn together under 

 the influence of their mutual attractions according to the 

 gravitational- molecular law that : — 



a. Whatever energy is lost is lost equally by the n bodies. 



b. Whatever energy is retained is retained equally by the 

 n bodies. 



c. When the bodies are close together, as in a liquid, the 

 average attraction on an interior particle is balanced, but 

 this balance of attraction is destroyed many times in a second. 

 Any molecule is continually under the action of enormously 

 great attractions. Its velocity must therefore be continually 

 changing — at one instant nearly zero, at another instant 

 probably very great. If its motion is viewed as a whole, a 

 molecule of a liquid probably proceeds with great velocity 

 around a more or less changing centre of mass. 



We might, in line with the last sentence of section 19, 

 consider temperature motion to be the motion of the centres 

 of mass suggested in c above. But these centres of mass 

 would usually be mere mathematical points only occasionally 

 coinciding with a real mass. A motion of these centres of 

 mass would therefore be in reality a motion of the individual 

 molecules above the motion required for maintaining their 

 equilibrium under the given attractive forces. Probably a 

 clearer idea of the true situation can be obtained from a 

 different standpoint. 



21. Consider a circular orbit and a corresponding ellip- 

 tical orbit. (By a corresponding orbit I mean an orbit that 

 could be substituted without adding to or subtracting from 

 the energy of the system.) The amount of energy retained 



